Fairness in multi-class multi-group classification problems via contextial coherent risk measures
Abstract
We propose a new design of fair classifiers for multi-class classification problems in the presence of vector-valued sensitive attributes. In that scenario each sensitive attribute has multiple values and forms several groups relevant to the fairness consideration. Naturally those groups are overlapping and one should also analyze the interaction of factors. Additionally, the decision makers aided by the classification should not violate individual rights at the expense of satisfying fairness metrics at the group level. We propose an approach using the theory and methods of coherent measures of risk aiming at resolving the fairness challenges. Further, we propose a specialized numerical method for solving the resulting optimization problem. The method scales well with the increase of the number of observations. Additionally, we note that the obtained classifier is robust with respect to corrupted data or to situation when data is scarce. We demonstrate the advantages of the proposed framework in comparison to the support-vector machine framework and other methods handling fairness.
Related
- Surrogate Losses for Fairness-Aware Classification: Fisher Consistency and Calibration
- A law of large numbers for Macdonald coherent measures
- A Classification of Fourier Summation Formulas and Crystalline Measures
- Sharp Brezis--Merle Inequalities for Polyharmonic Dirichlet Problems: Extremal Measures and Critical Asymptotics
- The Role of Uncertainty in Assessing the Fairness of Machine Learning Models
BibTeX
@misc{dentcheva2026fairness,
title = {{Fairness in multi-class multi-group classification problems via contextial coherent risk measures}},
author = {Darinka Dentcheva and Xiangyu Tian},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.30223},
eprint = {2608.30223},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.30223},
note = {CC BY-SA 4.0},
}
RIS
TY - UNPB
TI - Fairness in multi-class multi-group classification problems via contextial coherent risk measures
AU - Darinka Dentcheva
AU - Xiangyu Tian
PY - 2026
DA - 2026/08/31
PB - arXiv
DO - 10.48550/arxiv.2608.30223
UR - https://arxiv.org/abs/2608.30223
AB - We propose a new design of fair classifiers for multi-class classification problems in the presence of vector-valued sensitive attributes. In that scenario each sensitive attribute has multiple values and forms several groups relevant to the fairness consideration. Naturally those groups are overlapping and one should also analyze the interaction of factors. Additionally, the decision makers aided by the classification should not violate individual rights at the expense of satisfying fairness metrics at the group level. We propose an approach using the theory and methods of coherent measures of risk aiming at resolving the fairness challenges. Further, we propose a specialized numerical method for solving the resulting optimization problem. The method scales well with the increase of the number of observations. Additionally, we note that the obtained classifier is robust with respect to corrupted data or to situation when data is scarce. We demonstrate the advantages of the proposed framework in comparison to the support-vector machine framework and other methods handling fairness.
ER -